Microgrid lines

[vtnate]

Any background context you want to provide?

The ACLine.mot template for microgrid power lines doesn't handle the wires properly. This fixes it so the template builds a model the way I expect.

What does this PR accomplish?

• Update microgrid sys-param test file with modern output from opendss
• Fixes bugs in build-sys-param code that crashed if reopt data wasn't available
• Maps from OpenDSS output to MBL wires
• Adds wires to the line properly

How should this be manually tested?

1. poetry run pytest tests/GMT_Lib/test_gmt_lib.py -m 'not simulation'
2. Load Dymola with MSLv4, MBLv9, and tests/GMT_Lib/output/DistributionLines/ACLine*.mo
3. Simulate the model in Dymola
4. Inspect output of simulation

What are the relevant tickets?

Resolves #466

[amyeallen1]

Hi Nate,
I tried out the test, and it works perfectly to generate the model package, which then simulates without errors. I simulated the model, and also the ACLineMedium model from the buildings library, and there were some elements of the results that seemed unexpected to me, but they were the same in both cases. It's possible that I'm misinterpreting, or using different model inputs would generate more expected results. Here were those two items that I noted about the model results , for both this model and the original one in the MBL, (which I don't think have anything to do with your PR, so this is tangential):

1. There's no power dissipated thru any of the resistors, despite there being current thru the line, and a non-zero resistance value assigned.
   

2. The phase angle vector (in radians, according to the documentation) appears to be being reported cumulatively, and as a scalar:
   

Nate and Nick, do either of you have any ideas on what might explain these two things? Thanks! @vtnate @nllong

[nllong]

Hi Nate, I tried out the test, and it works perfectly to generate the model package, which then simulates without errors. I simulated the model, and also the ACLineMedium model from the buildings library, and there were some elements of the results that seemed unexpected to me, but they were the same in both cases. It's possible that I'm misinterpreting, or using different model inputs would generate more expected results. Here were those two items that I noted about the model results , for both this model and the original one in the MBL, (which I don't think have anything to do with your PR, so this is tangential):

1. There's no power dissipated thru any of the resistors, despite there being current thru the line, and a non-zero resistance value assigned.
   
2. The phase angle vector (in radians, according to the documentation) appears to be being reported cumulatively, and as a scalar:
   

Nate and Nick, do either of you have any ideas on what might explain these two things? Thanks! @vtnate @nllong

Can you plot the current and voltage too? This does seem strange, but it is good that MBL and GMT are resulting in the same. The cumulative phase angle is also strange... should be instantaneous.

[amyeallen1]

Hi Nick, and Nate, @nllong @vtnate
Thanks for considering this. Here are plots of current at port n (the other port is generally equal and opposite). These two lines are representative of the others.

These were representative of the voltage values in the lines, with the other element of the voltage vector at the node being zero.

Also, there seem to be some slight variations in current values between the GMT model (the higher value shown here) and the underlying MBL model, that aren't explained by different line numbering. I haven't id'd any explanation for that as yet. \

[amyeallen1]

Based on Mingzhe's comments, it appears that the model performed as expected. : )